TrueSource

Name: TrueSource
Author: agentpit-io

作者：agentpit-io版本 0.1.1i★ –i在你的电脑上运行

注册名称io.github.hangeaiagent/truesource-mcp

安装

概览

作者提供的说明（英文原文）

First-hand market signals: filings, procurement wins, R&D expansion, macro data

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数：意见、测量和人气是不同的东西。

用户评价i暂无评价评价功能将在后续阶段与账户功能一起开放
我们的基础检查i1 项中通过 0 项依据公开信息检查；尚未扫描代码
使用情况i尚未统计★ –

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

连接方式: 在你的电脑上运行
传输方式: stdio
分类: 其他
你需要填写：HUNTER_API_KEY

安装

选择你使用的 AI 应用，然后复制设置。

安装之前

基础检查：1 项中通过 0 项
在你的电脑上运行，拥有你的用户账户的权限（文件、网络、应用）。
需要你自己填写： HUNTER_API_KEY, TRUESOURCE_URL, TRUESOURCE_MAX_ITEMS

标为机密的值是你账户的钥匙。千万不要分享给别人，也不要粘贴到聊天里。

你需要

PyPI已安装 Python 的 uv（用于 uvx 命令）

你用的是哪个应用？

ClaudeClaude CodeCursorVS CodeChatGPT

在上方选择你的应用。我们会记住它，下次按钮会显示“安装到 …”。

打开 Claude Desktop → Settings → Developer → Edit Config，把下面的内容粘贴到 claude_desktop_config.json

{
  "mcpServers": {
    "truesource": {
      "command": "uvx",
      "args": [
        "truesource-mcp"
      ],
      "env": {
        "HUNTER_API_KEY": "your-value",
        "TRUESOURCE_URL": "your-value",
        "TRUESOURCE_MAX_ITEMS": "your-value"
      }
    }
  }
}

在终端中运行这条命令

claude mcp add truesource --env HUNTER_API_KEY=your-value --env TRUESOURCE_URL=your-value --env TRUESOURCE_MAX_ITEMS=your-value -- uvx truesource-mcp

运行前请先填写空着的值。

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json，或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "truesource": {
      "command": "uvx",
      "args": [
        "truesource-mcp"
      ],
      "env": {
        "HUNTER_API_KEY": "your-value",
        "TRUESOURCE_URL": "your-value",
        "TRUESOURCE_MAX_ITEMS": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用，然后应用会显示设置并请你确认。密钥的值会留空，由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "truesource": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "truesource-mcp"
      ],
      "env": {
        "HUNTER_API_KEY": "your-value",
        "TRUESOURCE_URL": "your-value",
        "TRUESOURCE_MAX_ITEMS": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用，然后应用会显示设置并请你确认。密钥的值会留空，由你自己填写。

这个服务器在你的电脑上运行，而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。

你的选择通过 Cookie 保存在这台设备上，无需账户。

需要你填写的值

HUNTER_API_KEY 机密必填 env
API key from https://hunter.agentpit.io/dev/api-keys
TRUESOURCE_URL env
Override the upstream endpoint if you self-host
TRUESOURCE_MAX_ITEMS env
Cap on items per response

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同，请以 README 为准。

它能做什么

它在你的电脑上运行；安装后，工具列表会显示在你的 AI 应用中。

基础检查i

我们检查的是任何人都能看到的信息；我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月8日

! 注意
作者与仓库一致
发布者（hangeaiagent）不是仓库所有者（agentpit-io）；可能是其他人代为上架的
… 尚未检查
仍在维护
我们还没有从 GitHub 读取这个仓库
… 尚未检查
有许可证
我们还没有从 GitHub 读取这个仓库
– 不适用
端点有响应
它在你的电脑上运行，没有互联网端点
– 不适用
使用 OAuth 登录
在你的电脑上运行，改用本地设置
… 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行；我们只能读取互联网端点的工具列表

与“其他”中的其他服务器比较i

星标数多于此分类中其他 18437 个服务器的 0%
发布到注册中心的时间比此分类中其他 18437 个服务器的 41% 更新
基础检查结果等于或优于此分类中其他 18437 个服务器的 21%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时，星标数才计入。

常见问题

TrueSource 安全吗？

它在 1 项适用的基础检查中通过了 0 项。我们没有阅读代码。在添加密钥或令牌之前，请先看看这个服务器要求哪些权限，并使用只具备所需权限的密钥。

它需要哪些密钥或代码？

你需要填写：HUNTER_API_KEY

它在我的电脑上运行还是在云端？

它在你的电脑上运行。你的 AI 应用会在需要时启动它，数据不会经过作者的服务器（除非它自己调用在线服务）。

它还在维护吗？

GitHub 仓库最后更新于–，最新版本（0.1.1）于2个月前进入注册中心。

哪些应用可以使用它？

支持 MCP 的应用，例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Python (uv)。

有什么可以替代它？

同一分类中有：Spec Driven Development, Remote Desktop Commander, ProAgentStore

替代选择

同一分类中其他作者的服务器，按星标数排序。

如何看卡片：★ GitHub 星标i三大支柱：评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Spec Driven Development

作者：formulahendry

MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.

在你的电脑上运行★ 440MIT14个月前更新

安装

Remote Desktop Commander

作者：desktop-commander

Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.

在云端运行★ 2113天前更新

安装

ProAgentStore

作者：ProAgentStore

MCP-first control plane for ProAgentStore agents and private instances.

在云端运行今天更新

安装

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

truesource-mcp

First-hand market signals your AI agent can actually cite.

An MCP server wrapping TrueSource — a crawler and AI-retrieval stack that watches things which have *already happened* and carry a verifiable source: exchange filings, government procurement awards, R&D expansion moves, northbound holdings, and official macro releases.

It does not summarise analyst reports and it does not emit buy/sell calls. Every signal comes with a date and where it came from.

Tools

Tool	What it does	Speed
`truesource_procurement(days, limit)`	Recent government procurement awards — who is actually winning contracts	fast
`truesource_macro(days)`	Macro releases from the statistics bureau, customs and industry bodies	fast
`truesource_daily_brief(symbols)`	Signal digest + alert level for a batch of tickers (last 3 days)	fast
`truesource_alert_signals(symbols)`	Only signals that crossed an alert threshold (last 26 hours)	fast
`truesource_report(symbol)`	Full research report — 35 pre-built AI-compute names only	fast
`truesource_scout(symbol, name)`	Live full collection for any A-share	30–60s

Procurement is the most distinctive one: a contract award is a fact that has already occurred, which is a harder input than a forecast.

Quick start

You need an API key first — see below.

Claude Desktop / Cursor

Add to claude_desktop_config.json:

"mcpServers": { "truesource": { "command": "uvx", "args": ["truesource-mcp"], "env": { "HUNTER_API_KEY": "hunt_tools_xxx" } } } }

Command line

# no install needed

HUNTER_API_KEY=hunt_tools_xxx uvx truesource-mcp

or install it

pip install truesource-mcp HUNTER_API_KEY=hunt_tools_xxx truesource-mcp

API key

This server has no keyless mode, and cannot have one — the upstream gateway returns 403 without a key.

The reason is cost: truesource_scout runs real crawlers plus a Gemini search on every call, so each invocation spends money on third-party APIs. Anonymous access would be trivially expensive to abuse and impossible to attribute.

Get one at https://hunter.agentpit.io/dev/api-keys — sign in, click 申请 API Key, copy the hunt_tools_… key (shown once), set it as HUNTER_API_KEY.

If the key is missing, this server does not return sample data or a fake success. It returns an error that tells you exactly where to get a key.

Environment variables

Variable	Default	Notes
`HUNTER_API_KEY`	—	Required. `hunt_tools_` key
`TRUESOURCE_URL`	`https://hunter.agentpit.io/api/saas/truesource`	Point at your own deployment if you self-host
`TRUESOURCE_TIMEOUT`	`20`	Seconds. `scout` gets its own 120s floor regardless
`TRUESOURCE_MAX_ITEMS`	`40`	Cap on items per response
`TRUESOURCE_MCP_TRANSPORT`	`stdio`	`stdio` \	`streamable-http` \	`sse`
`TRUESOURCE_MCP_HOST` / `_PORT`	`0.0.0.0` / `8933`	Remote transports only

stdio is what Claude Desktop, Cursor and uvx use. Run a remote transport only if one long-running server should serve several clients — and prefer streamable-http, since the MCP spec marks SSE as deprecated.

Three things that will bite you otherwise

`grey` is not `green`. In daily_brief, an alert level of grey means *no signal was collected for this ticker in the last 3 days*. That may mean the stock was quiet, or it may mean coverage missed it. It does not mean "safe".

An empty result is a conclusion, not a failure. alert_signals returning nothing means these tickers crossed no threshold in the last 26 hours. That is information. This server never dresses an empty result up as an error, and never fills it with placeholder rows.

Truncation is always disclosed. Responses are capped at TRUESOURCE_MAX_ITEMS (40 by default). When that bites, the payload carries truncated: true, the real total, and a note not to draw conclusions from the visible slice. Raise the cap or narrow the date range instead.

truesource_report also will not silently fall back to scout when a ticker isn't one of the 35 pre-built names. Scout costs 30–60 seconds and real money — that choice should be made deliberately, not happen behind your back.

Errors you might hit

Error	Meaning
`missing_api_key`	`HUNTER_API_KEY` not set — see API key
`invalid_api_key` (401)	Key wrong, incomplete or revoked
`forbidden` (403)	Key valid but lacks access to this endpoint
`not_found` (404)	For `report`: ticker isn't a pre-built name — use `truesource_scout`
`timeout`	`scout` genuinely takes 30–60s; raise `TRUESOURCE_TIMEOUT`
`upstream_down` (502/503)	Crawler service is down; retry later

[kronos-mcp](https://github.com/agentpit-io/hunter-community/tree/main/tools/kronos-mcp) — K-line forecasting for A-shares, US and HK equities
[akshare-mcp](https://github.com/agentpit-io/hunter-community/tree/main/tools/akshare-mcp) — lets a model explore AKShare's 1000+ China market data functions on its own
[hunter-community](https://github.com/agentpit-io/hunter-community) — the open-source stack these come from

Apache-2.0.